Survival analysis models are used primarily in medicine and churn analysis. Due to many applications, we are witnessing a fast development of a wide range of black-box survival models. Their lack of interpretability makes them unusable for analyzes that require an understanding of the model behavior.

An R package survxai is a tool for creating explanations of survival models. For both, complex and simple survival models. It also enables to compare them. Currently, four explanation methods are implemented. We can divide them into 2 groups: local and global.



Informations about changes in survxai relases can be found in NEWS.

How to contribute

Informations about creating issues and pull requests in survxai can be found in CONTRIBUTING.


Work on this package is financially supported by the ‘NCN Opus grant 2016/21/B/ST6/02176’.